6

I finished my master thesis and my advisor asked me if I would like to publish a paper on it. Being my first paper, I was very excited when he first told me, so I immediately started reading the journal rules about how to write it properly.

Now I am at the point in which I should start writing, but I actually do not know how to organize things. I asked my advisor and he said it should be a sort of summarisation of the thesis, but I fear some contents of the thesis, that were new to me at the beginning, might be taken for granted from research people and so I will actually write obvious things.

Also, I have some contents which, I think, are quite interesting in the sense that I saw only one other paper using some of these concept, but without providing any information about the set-up of the code, which is something I would like to add since I feel it might be useful to have an example on how to do it, especially if you are at the beginning as I was when I started the thesis. Still, the idea is not new. So, how to deal with this? My feeling at the moment is that the paper does not worth the effort.

  • How much of the background stuff to include might depend on your field. What field are you in? – Tobias Kildetoft Apr 20 '15 at 7:38
  • 1
    Aerospace engineering – Rhei Apr 20 '15 at 7:45
  • It is a good idea to edit that into the question for better visibility. I am not familiar with your field, but in my field (math), I would advise that you only include those parts of the thesis that are actually new knowledge, together with sufficient background for those things to make sense. Better a shorter paper that is more to the point than one that gets rejected for containing too much "trivial" stuff. A possible exception to this can be including specific examples though. – Tobias Kildetoft Apr 20 '15 at 7:52
  • 2
    You are a master's student. Your advisor presumably will help you edit your paper. Write your paper up, err on the side of including more detail, not less. Show it to your advisor. If you are including stuff that is too trivial, he will surely tell you. – Willie Wong Apr 20 '15 at 8:17
7

You could structure your material as follows. To my perception (at least in computer science) this is more or less standard structure.

  1. Introduction: The bigger picture in which your approach fits in: the problem you are trying to solve. A short description that guides the reader from the more general to the more specific issue.
  2. You could have an extra section regarding some preliminary concepts that are crucial to understanding your work. The nice thing with journals is that you do not have the space restrictions that you have in conferences, and thus you can expand a little.
  3. Related work: Attempts that already exist in the literature, and how you compare with them. Why your solution would be an advance in the state-of-the-art, and under which conditions.
  4. Description of your work: as detailed and as formal you can be regarding e.g. a system architecture, an algorithm, the core idea of the paper.
  5. Performance measurements. A comparison to "competing" approaches in terms of performance.
  6. Optionally, a discussion section, where you discuss about your experiments. This could be merged in the previous section.
  7. Conclusions and future work. A section where you provide your most important findings and give directions towards which your work can be expanded.

Of course, the above are merely indicative and do not apply to all cases. Personally, this is how would start drafting a new journal publication.

My feeling at the moment is that the paper does not worth the effort.

It is certainly worth the effort. Why not? Even if the idea is not new, your paper would be a study on how your approach performs in certain conditions.

Funny as it may be, this presentation gives some quality advice: http://research.microsoft.com/en-us/um/people/simonpj/papers/giving-a-talk/writing-a-paper-slides.pdf

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.